Source : INDIA TODAY NEWS
Artificial Intelligence is changing how software is built. AI agents are now writing code much faster than human developers can read and review it. To solve this problem, former GitHub CEO Thomas Dohmke has launched a new platform called Entire.
Announced just two days before GitHub’s annual Universe conference, Entire is a brand-new service for hosting, reviewing, and shipping code. It is designed specifically for a world where AI does the heavy lifting, positioning itself as a direct rival to GitHub.
advertisement
Why do we need a new platform?
Currently, platforms like GitHub are struggling to handle the massive workload created by AI agents. In September alone, developers and AI tools saved 7.38 billion code changes on GitHub. This heavy load has caused several major outages recently, including an almost eight-hour blackout in August.
Dohmke, who left GitHub last year, realized that traditional platforms were not built for this speed. His new startup, which currently has a fully remote team of 50 employees, promises a network that can handle as many code pushes per second as GitHub, with AI agents pushing code all at the same time without crashing.
Goodbye Pull Requests, Hello ‘Trails’
In traditional coding, developers use something called a “pull request” to propose changes, which humans then have to read line-by-line. Entire is completely removing this process and replacing it with a feature called “Trails.”
With Trails, when new code is submitted, multiple AI agents automatically evaluate it and give it a score. Instead of reading every single line of code, human developers just look at a dashboard of these scores. The code only goes live if it gets a green light from the AI, passes automated tests, and gets a final human sign-off.
“Code review was always a bottleneck,” Dohmke said at a Madrona event last month. “I’ve never worked in a software company where I didn’t wait for somebody else to review my pull request… And now that bottleneck is 10 times bigger.”
Continuous Memory for AI
Another standout feature is “Continuous Memory.” When AI agents write code on Entire, the platform remembers the context behind it. If a new AI agent comes in later, it can search past coding sessions to understand exactly why a specific change was made.
According to the company, this smart memory map also helps cut down the AI’s operating costs by up to 71%.
Lightning-fast speeds in India
Entire claims to be incredibly fast. Based on the company’s own testing, it can process code pushes up to 89 times faster than GitHub.
To make sure developers get a smooth, lag-free experience, Entire is hosting private code repositories across multiple global regions, including the U.S., Europe, Australia, and India. This means Indian developers and their AI tools will get fast and reliable access to their projects.
Pricing and backing
The Entire platform itself is free to use right now. Developers only have to pay for the AI model “tokens” used during the automated code reviews. Reportedly, the company is offering a $100 starter credit to help users try it out.
Interestingly, while Entire is directly competing with the Microsoft-owned GitHub, Microsoft’s own venture arm, M12, is actually one of Entire’s investors. The startup raised $60 million in seed funding earlier this year at a $300 million valuation, with backing from major investors like Gergely Orosz, Olivier Pomel, and Yahoo co-founder Jerry Yang.
– Ends
SOURCE :- TIMES OF INDIA
